Open Access

Table 4

Comparison of classification results under different levels of Gaussian noise and salt-and-pepper noise by different models.

Reference Model Add noise type Add noise level Accuracy
This work CvT-13 Gaussian noise sigma=0 0.988
sigma=5 0.931
sigma=25 0.864
sigma=50 0.837

Salt-and-pepper noise amount=0 0.988
amount=0.05 0.902
amount=0.1 0.831
amount=0.2 0.815

[1] EfflcientNet-B5 Gaussian noise sigma=0 0.968
sigma=5 0.783
sigma=20 0.847
sigma=50 0.815

References. [1] Ai et al. (2022).

Current usage metrics show cumulative count of Article Views (full-text article views including HTML views, PDF and ePub downloads, according to the available data) and Abstracts Views on Vision4Press platform.

Data correspond to usage on the plateform after 2015. The current usage metrics is available 48-96 hours after online publication and is updated daily on week days.

Initial download of the metrics may take a while.